Findings From ALTS

Abstract

Objective: To summarize some of the main results of the ASCUS/LSIL Triage Study (ALTS), a randomized multicenter clinical trial to compare management strategies for women with ASCUS and LSIL cytologic findings. Study Design: A total of 5060 women, 3488 with an ASCUS (atypical squamous cells of undetermined significance) and 1572 with LSIL (low-grade squamous intraepithelial lesion) cytology, were randomly assigned to 1 of 3 arms: (1) immediate colposcopy regardless of enrollment test results; (2) testing for human papillomavirus (HPV) DNA with referral to colposcopy for a positive result; and (3) conservative management based on repeat cytology with HSIL (high-grade squamous intraepithelial lesion) as the threshold for referral to colposcopy. All arms included 2 years of semiannual follow-up and colposcopy at exit for patient safety and disease ascertainment. Results: For women with ASCUS, the cumulative detection of CIN3 and cancer (CIN3+) was 8–9% over 2 years of follow-up and did not vary significantly by study arm. A single HPV test at enrollment was positive in 92% of the cases of CIN3+ and was positive in 53% of women overall. Two repeat cytology evaluations, with a low referral threshold of ASCUS+, demonstrated high sensitivity of 95% for CIN3+ but would result in 67% of women sent to colposcopy. The strategy of immediate colposcopy for all women detected only about half of the cumulative (2-year) cases of CIN3+. For women with LSIL, the cumulative detection of CIN3+ was 15.0% and did not vary significantly by arm. Of note, the detection of CIN3+ for LSIL was similar to the risk for HPV-positive ASCUS, 15.2%, supporting a unified view of the 2 interpretations. No useful triage test strategy was identified for LSIL, which was predominantly HPV positive (83%) by Hybrid Capture 2 (HC2). Some LSIL was associated with nononcogenic HPV types by PCR testing. Only 5–10% of LSIL was HPV negative by both HC2 and PCR; these cases included instances of false-negative HPV test results or false-positive cytologic interpretations of LSIL. ALTS analyses have confirmed the 1.0 RLU/PC positive cutpoint for HC2 as the most appropriate for clinical utility. HC2 was more reproducible with an overall κ of 0.84 compared to liquid based cytology interpretation and histologic diagnosis of colposcopically directed biopsies, both of which demonstrated a κ of 0.46. Conclusion: The ALTS findings have informed the development of evidence-based management guidelines for women with ASCUS and LSIL cytology. HPV triage is a cost-effective management approach for women with ASCUS and spares approximately half of women the need for a colposcopic examination. Repeat cytology requires multiple follow-up vists to achieve similar sensitivity as HPV testing. There was no effective triage identified for women with LSIL cytology; consequently guidelines recommend evaluation by colposcopy. The prospective follow-up of ALTS participants provided the opportunity to identify missed prevalent disease and demonstrated that no one test was completely sensitive for detection of CIN3+. All tests, including colposcopically directed biopsy (previously considered the “gold standard”), have limitations.